The prime objective is to determine the duration and characteristics of naturally and artificially acquired immunity to influenza by longitudinal studies in three population cohorts under clinical and serological observation since the late 1960's. Antibody response to nucleoprotein, hemagglutinin and neuraminidase antigens of influenza A virus will be measured, and correlated with incidence of influenza infection and illness. New methods for assay of anti-hemagglutinin, such as single radial hemolysis, applications of immunodiffusion, rocket immunoelectrophoresis, and radio-immunoassay will be investigated and evaluated. The humoral antibody response and infection rates will be studied by age and in family constellations. Antibody decay will be measured, evidence of reinfection in successive epidemics and evidence for protective antibody will be sought, and responses and infection in those previously vaccinated will be analyzed. The hypothesis that antibody to neuraminidase protects against disease and prevents transmission but does not necessarily prevent infection, will be tested. The possibility of long-term adverse effects of influenza vaccine will be investigated. The antibody response in patients who develop pneumonia as a complication to influenza will be compared with an age and sex matched sample of persons who had uncomplicated influenza infection with and without report of typical influenza illness. Knowledge of the type of antibody present in persons seemingly immune to influenza is essential in predicting the effectiveness and guiding the use of vaccines. Most sera have already been collected, and additional collections are aided by another grant.